The latest version of NMIS can be downloaded from the Opmantek.com Download Page
Table of Contents |
---|
NMIS 8.
...
7.
...
1G
This is a maintenance release of NMIS , which was published on 17 Aug 201829 June 2020.
Highlights for the 8.
...
7.
...
1G release
...
- Fix a polling bug that makes a node be found as candidate for WMI polling.
- Model fixes and updates.
We highly recommend to upgrade to this version from NMIS 8.7.0G.
NMIS 8.7.0G
This release of NMIS was published on 22 May 2020.
Highlights for the 8.7.0G release
8.7.0G is a major release with lots of new features, improvements and bug fixes.
Installer fixes
- Get PAM module from repositories
- Now installer uses cpanm for cpan installs as it is more flexible and does not need initial configuration as cpan does
Model Enhancements
- support for new models:
- Viptella models
- Ericsson Router
- Vyatta VYOS devices
- Cisco BGP MIB
- Huawei HQOS graphs
- Stats metrics to the Cisco routing model
- Bluecoat Model
- f5 model
- Model Improvements:
- Cisco Meraki and Meraki Cloud
- Improved models for Juniper IVE devices
- Improved Fortinet FG model
- Improved Microtik model
- Improved Cisco Nexus model
- Display temperature of the sensors in AristaSwitch model
- Fixes for the Cisco Temperature model. Updates to the Cisco models to include temperature values
- Work around net-snmp snmpd which reports almost anything as hrfixeddisk
- Removing the degrees symbol from Units in Alerts
- Added 'option' => 'small' to generate small graph when small graph requested in maxbits
- Some graphs updated to show Max values, not only averages
- Enhancing the Host Resource Model with thresholding
- Added Opengear to Model.nmis to use Default-HC
- Added lldpLocal to Model-Policy
- Adding the missing ADSL graph to the NMIS source and fix ADSL model files
- Added new RouteNumber section so variable accessible through stats
- Common-database.nmis updated to include 'TeldatQoSStat' and 'TeldatBTSStat' entries
Plugin Improvements
- Cisco Meraki Cloud
- Viptella models
- Added the OSS banner to the plugin code
- Improvement in NMIS having a plugin include reach data
- Enhancements to NMIS for health metrics for plugins and implementing ifNodeCollectable
Collect Improvements
- Only run services polling on nodes which have service polling defined
- Improvement to ifAdminStatus change detection to handle ifIndex changes. New event (Interface IfAdminStatus Changed)
- Adding 300 seconds to node reset detection for odd Linux SNMP sysUpTime
- Added collect_wmi/snmp properties to allow disabling either protocol (snmp_enabled, wmi_enabled). Also, added the migration script migrate_collectpatch to support this properties, added support for SNMPv3. Added an script test in admin/upgrade_node_collect
- Optimisation to stop running services if no service polling required on a node
- Improved debugging for a single node, runEscalate is skipped for a single node collect
- Add a new configuration option sort_due_nodes, to order nodes based on last_time_poll
- Improving the polling so that nodes do not get polled too frequently
- Added escalate_poll_cycle to control when the escalate cycle is run
Admin Tools
- Added new utility script admin/import_zenoss_backup to import devices from Zenoss device dump file (Example in admin/samples/zenbatchdump_sample.txt )
- model_discovery to automatically determine which existing modelling exists
- alcatel_interface_discovery admin tool has been deprecated
Other Fixes and Improvements
- Updated printCrontab to split the running of collect and services to separate cycles. Also include additional info for run=escalate
- fix in node.pl to show type=metrics, group=network graphs
- Added support to preserve -- when lowercase names are wanted
- Improve the debug output. Remove passwords from logs
- Added support for Common-event.nmis event model policy handling for custom node roles
- Added Node Count per poll cycle count
- Added handling for new types of loopback interfaces called lo0
- The minimum length of time that a graph can show has been reduced from 30 min to 10 min, this allows to see shorter periods with better details
- Updated polling summary to include demoted nodes and pingonly nodes, show stats for demoted and non snmp polled nodes
- CVAR support for threshold controls
- Add interface description in custom test alerts
- Fix issue with notifications for elements with /
- Update the installed modules detection
- Added support for Crowd Auth
- Add support for env_temp with the new configuration item env_temp_legacy_support
NMIS8 installer may stop at CPAN attempt to install WWW::Mechanize on Redhat6 or Centos6
Should the install of NMIS8 stop at error
- t/local/back.t ........................... 1/47 Error GETing http://[::1]:44662/: Can't connect to ::1:44662 (Bad hostname '::1') at t/local/back.t line 53.
please install WWW::Mechanize using the following command then re-attempt install of NMIS8:
Code Block |
---|
cpan notest install WWW::Mechanize
# or if the above doesn't work, type cpan hit ENTER KEY and then from with cpan process then exit the cpan process:
notest install WWW::Mechanize
exit |
NMIS 8.6.8G
This maintenance release of NMIS was published on 10 Jul 2019.
Highlights for the 8.6.8G release
- Various modelling improvements
updates for Arista and Arris devices, new models for Dell switches, updated and extended modelling of the Host Resources MIB (with plugin),
adjustments for Cisco Catalyst IOSXE devices, Nexus CBQoS modelling, substantially improved modelling for Net-SNMP devices. - The node selection, scheduling of operations and process management have been reworked extensively, and NMIS spreads the load now better and more evenly.
- the new config options
nmis_max_nodes_per_collect_cycle
,nmis_max_nodes_per_services_cycle
andnmis_max_nodes_per_update_cycle
were added to optionally limit how many nodes should be considered at most for an operation cycle. - the new commandline option
abort_after=N
instructs NMIS to abort unstarted collect operations for candidate nodes after N seconds (default: 60).
Any candidate nodes that remain unhandled that long after start up are skipped (and the fact is logged), so that the next NMIS invocation can safely pick candidate nodes without overlap.
- the new config options
- Misbehaving Service test programs are now handled more consistently.
All exit codes outside of the required range of 0-100 are logged and interpreted as the service being down. - The webtest example Service test program now supports sending custom headers using -H.
- some minor GUI improvements related to node editing
- JSON event logging was extended
- NMIS' data export now includes both average and max data from rrd.
- The NMIS selftest is now less likely to produce false positives and runs faster.
- The default data retention periods for historic events has been reduced to 7 days.
- Various installer improvements, improved robustness and reduced number of repeat questions
- A new stateless event named 'Interface ifAdminStatus Changed' is raised when an interface's administrative status changes
- The installer now handles Debian 10 (Buster), and NMIS works correctly on Debian 8 to 10.
- Various minor adjustments and fixes for the NMIS graphs for better display with recent rrdtool versions
- Robustness improvements for WMI, for pinging with restrictive firewalls and for backup polling
- NMIS now supports Authentication using the host's PAM infrastructure.
- Escalation emails for failing or restored services now include the service's Description if present.
- Various other bug fixes and improvements
NMIS 8.6.7G
This is a maintenance release of NMIS, which was published on 17 Aug 2018.
Highlights for the 8.6.7G release
- The node dashboard widgets and graph drill-ins now load substantially faster,
and graph files are now cached for up to 60 seconds (configurable viagraph_file_maxage
config item). - SNMP Uptime counter wrapping (at 497 days) is now handled more robustly.
- The logic for group display/exclusion in the GUI widgets was repaired,
and now only nodes belonging to the configured groups (ingroup_list
) are displayed.
(Like before, all active nodes are handled by the polling and event engine, regardless of group membership.) - The handling of node resets and long-term unpollable nodes is now more robust and reliable.
- The Polling Failover feature was extended to provide more fine-grained events and node states.
- NMIS now provides a configurable rule for acceptable node names (using the config item
node_name_rule
) which is consistely applied everywhere;
the shipped default rule was also relaxed and now allows letters, numbers, space, underscore, hyphen and dot. - NMIS now supports WMI polling in windows domains more robustly and username and domain can be provided in multiple formats.
- Various modelling improvements, e.g. the
nocollect
directive is now supported in systemHealth model sections. - New and improved devices support for Nexus 7k, Adtran, Cisco, Arista, Mikrotik, Netvanta, PulseSecure, Juniper.
admin/node_admin.pl
now supports bulk export and import..- Various robustness improvements, e.g. in
fpingd.pl
, more fine-grained logging, more flexible and precise polling candidate timing decisions.
NMIS 8.6.7G Scaling Tip
If you are installing NMIS 8.6.7G and you have a busy server, then it is important that you modify the cron entry for NMIS and change it.
The setting is found in /etc/cron.d/nmis and the default is this:
Code Block |
---|
* * * * * root /usr/local/nmis8/bin/nmis.pl type=collect mthread=true ; /usr/local/nmis8/bin/nmis.pl type=services mthread=true |
This should be changed to:
Code Block |
---|
*/1 * * * * root /usr/local/nmis8/bin/nmis.pl type=collect mthread=true
*/2 * * * * root /usr/local/nmis8/bin/nmis.pl type=services mthread=true |
It is important to note that the NMIS polling engine had an overhaul in NMIS 8.6.6 and NMIS 8.6.7 to improve how parallel threads were handled and that polling was kept up to date.
So when NMIS starts a poll every 1 minute, not all nodes will be polled, it will poll as many as it can in that time and then leave the others for the next poll cycle, this has the result of spreading the polling (and load on the server) out over 5 minutes. If you are not getting all nodes polled in 5 minutes, you will need more threads (and enough memory).
If you are polling many nodes every 1 minute, then you will need to size the server accordingly
...
.
NMIS 8.6.6G
This maintenance release of NMIS was published on 1 Jun 2018.
...
- The handling of nodes with large numbers of interfaces was substantially improved, and a number of optimisations to reduce the collect times for such devices were implemented.
- The event subsystem was reworked for improved performance, and a number of related bugs were identified and fixed.
- NMIS now uses locks for per-node operations to ensure correct sequencing even if an update or collect operation takes a very long time.
Time-exceeded events now also include better diagnostic info, i.e. the node name in question. - Improved Threshold Alerts for some measurements, e.g. CPU.
- New and improved models for Checkpoint, APC, ACME, Furukawa, Alcatel devices, Ubquiti, Microtek, Frogfoot, Cisco; better modelling of OSPF.
- A number of models have been reworked for greater modularity and ease of use. Many were adjusted to handle corner cases better.
- Updates of the Cisco product mibs.
- For newly added nodes a type=update operation is now run automatically if and when required.
- NMIS now detects problems with sending events via syslog better and logs them.
- Node Configuration overrides are now presented better in the GUI, and the nodeconf subsystem was reworked for better performance.
...
NMIS 8.5.8G
Friday 24 April 2015
...
Enhancements and Changes in 8.3.9G General Release
New and Improved Device Support
Added ESXi 5 SNMP support for host collection.
Added basic Nexus Support.
Added IP graphs in the models for certain devices
Removed Cisco specific content from the Generic and Default models.
Improved Threshold Granularity
Added more granularity to the default Threshold Model for a few policies, and added properties for control, to include group and roleType properties for even more granularity.
Config comparing and updating tools
admin/diffconfigs.pl /usr/local/nmis8/install/Config.nmis /usr/local/nmis8/conf/Config.nmis
admin/updateconfig.pl /usr/local/nmis8/install/Config.nmis /usr/local/nmis8/conf/Config.nmis
Display Company Logo on NMIS Dashboard
A "company_logo" option added to the Config.nmis, this can be any URL which will point to a logo to use, no bigger than 30px high is better.
Collect an interface based on description pattern
This overrides all other properties. This is added to the model in the interface section, looking something like this:
Code Block |
---|
'collect' => { 'Description' => '(CNOC|Collection)', }, |
The description is a regular expression which can contain complex patterns.
fpingd.pl Options
Added config option, 'fastping_node_poll' => '300' and 'fastping_sleep' => '60', these control how many nodes fpingd will ping at once
fpingd.pl will use DNS by default or use what the OS would like it to use, using Perl gethostbyname.
New IPSLA Configuration Options
IPSLA configuration options from the NMIS config.
Code Block |
---|
# collect interval time in seconds my $collect_time = $C->{ipsla_collect_time} ? $C->{ipsla_collect_time} : 60; # collect interval of buckets in seconds my $bucket_interval = $C->{ipsla_bucket_interval} ? $C->{ipsla_bucket_interval} : 180; # extra collect buckets for polling cycle my $extra_buckets = $C->{ipsla_extra_buckets} ? $C->{ipsla_extra_buckets} : 5; |
Interface Persistance for Node Configuration
Node Configuration now uses the ifDescr to index manual changes to the interfaces, so that changes will be persisted even if the device changes the ifIndexes on reboot (as Cisco devices often do). Also fixed ifSpeed to be shortened.
Improved SNMP Connectivity Handling
Fixed problems with SNMP polling when SNMP is down, optionally can continue polling SNMP if fails, good for unreliable/congested links.
'snmp_stop_polling_on_error' => 'true'
Disable NMIS Polling
Disable polling temporarily by using the global_collect set to false in the config.
Polling Performance Log
Polling Log for see what is being polled and written to RRD, configure with polling_log. To enable polling log a file must be configured in Config.nmis and the file must exist.
Many Other Enhancements
Added HostInfo to tools.pl
Added NMIS::Modules to allow transparent integration with NMIS Extensions
New config option for handling wrapping of node names. wrap_node_names defaults to false
Added new host diagnostics, nmis_polling_summary
added config option, nmis_summary_poll_cycle, when set to false, you can run nmis.pl type=summary to improve polling performance.
Added config option for widget_refresh_time, default will be 180, this is for very busy dashboards which were constantly refreshing data.
Added view tables to the Access configuration for Authorisation, not allowed for anonymous and some allowed for guest. New Access.nmis template
Added a new log file, auth.log, all Auth logging will be sent to that log file.
Added missing items back into Config for various things, like PING options.
IPSLAD off by default, need to install SNMP_Session
Added SMTP Authentication and SMTP over SSL SUPPORT.
Changed network metrics selector to be a pull down instead of a clickable link.
Added NMIS::uselib as a package to better handle including RRDTool, fixed the things including use lib for rrdtool
Added more granular thresholds for Gigabit interfaces, others could be added to be speed dependant as well.
Bought SNMP handling options into configuration, snmp_stop_polling_on_error, snmp_timeout, snmp_retries
Fixes in 8.3.9G General Release
Fixed thresholding so that it processed the other items listed in the models, it was only processing the first item in the list.
Fixed a bug with SNMP polling where exceptions where not being handled causing SNMP session corruption. Only an issue in networks with very high latency and packet loss.
Fixed the left/right/centre in the Top10 report.
Fixed a bug with blank events in the event log.
Fixed Log Targets for viewing logs in Widgets
Fixed Javascript namespace where forms where conflicting
Fixed bug with customising locations, where on changing an update would use sysLocation from SNMP, but collect would use custom location.
Working with MTR
Sticky bit needs to be set on MTR to work, owned by root and sticky
Code Block |
---|
chown root /usr/sbin/mtr chmod u+s /usr/sbin/mtr |
NMIS 8.3.1G
Thursday 1 December 2011
...